*Table 3, Replication Using Bootstrapped Standard Errors
*Originally Simmons and Danner, 2009 - Table 2

*Andrew Miller, MIT
*Alex Copulsky, MIT

*Table 2 demonstrating the effect for freedom house measure:
stcox  ratification iintensitycw truth iaid logcinc exec_a logmilper anymilitary if fh<2, robust cluster(country)

*Table 2, Model 1:

bootstrap, reps(1000) seed(123): stcox  ratification iintensitycw truth iaid logcinc exec_a logmilper anymilitary if polity>0, robust cluster(country)
bootstrap, reps(1000) seed(123): stcox  ratification iintensitycw truth iaid logcinc exec_a logmilper anymilitary if polity<1, robust cluster(country)

*Table 2, Model 2:

bootstrap, reps(1000) seed(123): stcox  ratification iintensitycw truth iaid logcinc exec_a logmilper anymilitary if irule>-1, robust cluster(country)
bootstrap, reps(1000) seed(123): stcox  ratification iintensitycw truth iaid logcinc exec_a logmilper anymilitary if irule<-1, robust cluster(country)

*Table 2, Model 3:

bootstrap, reps(1000) seed(123): stcox  ratification iintensitycw truth iaid logcinc exec_a logmilper anymilitary if polconiii>0, robust cluster(country)
bootstrap, reps(1000) seed(123): stcox  ratification iintensitycw truth iaid logcinc exec_a logmilper anymilitary if polconiii==0, robust cluster(country)
